3 Aug
2005
3 Aug
'05
6:30 p.m.
Subject: Re: Grrr - !%#*^@# Kaypro!
From: Fred Cisin <cisin at xenosoft.com>
Date: Wed, 03 Aug 2005 16:08:22 -0700 (PDT)
To: "General Discussion: On-Topic and Off-Topic Posts" <cctalk at
classiccmp.org>
On Wed, 3 Aug 2005, Allison wrote:
???! The value for R must be a valid sector number. For floppies
that is never zero.
Dave, your on track. The CHRN vlaues do not have
to corrospond
to actual head or cylinder. They do not even have to be in order
on the media! The only limits on them are R must be 1 or greater
(less than 255).
Which creates some complications with handling 128 byte sectors (R==0). My favorite format was sectors 1- 9 (512 DD) on
side one and
side two was 10-19 with H always set to 0.
Easily handled by 765, but not by Int13h. I have to be careful to differentiate which
complications are
due to the FDC chip, which to the FDC board/implementation,
and which are BIOS linitations.
Yep, The bios for the most part is a not a limiting factor if you program
around it. Its the board implmentations or the chips that replace
board implmentations that really are poor.
Nitpick: don't you mean 0 - 9 on first side? Or
did it actually have 9
sectors on one side and 10 sectors on the other?
;) That was why it was pretty wild. it was done for a joke in the lab
but no one could sort it out.
I've seen disks with 10h and 20h in the H field.
Yep, CHRN are supposed to carry Cylinder, Head, Record(sector) and
N=data length. But the allowable values are usually 0 or 1 to 255.
One engineer at Bell Labs used a 765 to make a nice block replaceable
DC300 tape system. He used R and H to form a 16 bit block number.
This was back in 1981 so it predates the PC based tape on floppy backup.
And, of course, there are some formats, such as
Superbrain,
where the data (sector content) is inverted,
but the headers (CHRN) are not.
that really bothered people for some reason.
Allison